Development of time-gated Raman proximal sensing for an earth observation platform

The Goldeneye project combines remote sensing and positioning technologies with proximal sensing to produce reference calibrated mineralogical maps through data fusion. The project brings together optical satellite sensor data, drone sensor aerial data both optical and electromagnetic as well as spectral ground sensor data. Satellite data can offer spectral signatures of large areas but suffers from limited spatial resolution and blind spots where the higher resolution satellite data is not available. Drone data can offer more variety in spectral wavelengths with higher resolution but there are some drawbacks as well. Namely, NIR vibrational spectroscopy requires background information for successful mineralogical analysis. In addition, the most interesting SWIR range is challenging due to large and very expensive sensors. To cope with these challenges of aerial data, proximal sensing can be applied in locations where satellite imagery is not available, and it can also produce reference information for the calibration of the spaceborne and airborne instruments. The conventionally used analyses for producing mineralogical information from field collected samples are the mineral liberation analysis (MLA) and X-Ray diffraction (XRD) which require extensive sample preparations and are laborious and slow. Goldeneye-project has studied the use of time-gated Raman for easier and more practical production of reference data at the field sites as well as from field collected rock samples. The benefit of Raman is an accurate characteristic spectral fingerprint and an ability to distinguish small mineralogical features as the detection spot is in the range of hundreds of microns. There are continuous wave Raman spectrometers, which are already field deployable. However, conventional Raman suffers from the auto-fluorescence emission triggered by the laser illumination, especially in light-colored rock samples. Time-gated (TG) Raman has the benefit of time-resolved sensing, where the Raman scattering is recorded before the fluorescence signal is over-powering the weaker scattered signal. TG-Raman can thus offer information from a wider variety of geological specimen than the conventional Raman. In Goldeneye-project, TG-Raman spectra were collected with custom sampling solution from mineral samples and drill cores from Erzgebirge exploration site in Germany. The data was analysed together with pXRF reference data to assess the benefit of the data fusion.

The paper has been prepared in the frame of the Horizon 2020 co-funded project GOLDENEYE, which has received funds through the Grant Agreement 869398.